Randomized response techniques for a multi-level attribute using a single sensitive question

Comparison of Estimators for Multi-level Randomized Response Data: Evidence from a Case of Sexual Identity

“…If it exists, the ML estimator (MLE) can be found in the parameter space 0 , 1 by means of constrained numerical procedures. Its MLE asymptotic property is found in Hsieh et al (2018).…”

“…Motivated by these contributions and the Bayesian approach’s benefits, we explore the Taiwan Social Change Survey (TSCS) as a case study in connection with the multi-level randomized response technique (MRRT) by Hsieh et al (2018). We compare the ML and Bayesian estimated proportions of individuals aged 18–54 years self-reporting as bisexual and homosexual among Taiwanese.…”

“…De Jong et al (2012) introduced a polytomous item RRT that increases reporting sexual proclivities related to DQ. Hsieh et al (2018) studied the percentages of respondents self-reporting as bisexual and homosexual in Taiwan. In many studies, where the method of moments or maximum likelihood (ML) method was used to analyze RRT data, the estimated population proportions were unacceptable, especially in surveys when the sample size or proportion of individuals bearing the sensitive attribute was very small.…”

“…The study is organized as follows. In the next section, we discuss MRRT for self-report of sexual identity from the 2012 TSCS and reveal the limitations of Hsieh et al (2018)'s ML method. In the section after that, we implement a Bayesian framework that overcomes the weaknesses of ML estimation of sensitive traits.…”

A Two-stage Multilevel Randomized Response Technique With Proportional Odds Models and Missing Covariates

“…Using this RR technique, the respondents are divided into two mutually exclusive groups by means of a randomization device, such as a spinner, dice, playing cards, random numbers, or a computer, to enable the respondents to answer the sensitive questions without revealing their true status about the stigmatized or private attribute to the interviewer. Since then, different developments and variants of the RR technique of Warner (1965) have been proposed by different authors (see, e.g., Christofides 2003; Franklin 1989; Greenberg et al 1969; Hsieh, Lee, and Tu 2018; Mangat 1994; Moors 1971; Lensvelt-Mulders et al 2005). Among them, Lensvelt-Mulders et al (2005) discussed two meta-analyses on 32 comparative RR technique studies in which the RR technique may yield a more reliable estimate compared with self-administered questionnaires or face-to-face interviews, especially when dealing with highly sensitive topics.…”